Rebuild logic between partial missing observations and complete missing observation (free run case), and fixed some edge cases for the wishart Q. #3267
Rebuild logic between partial missing observations and complete missing observation (free run case), and fixed some edge cases for the wishart Q.
Description
The free run is different than a normal run with partial missing observations for specific sets of sites. For the free run, we will estimate all state variables based on previous states and the potential Q from outside. For the normal run with partial missing observations, we will estimate state variables that match up with maximum types of observations across all sites. For the Wishart Q, some edge cases such as single site with single observation will be treated as Normal-Gamma conjugacy instead of MVN-Wishart conjugacy.
Motivation and Context
Review Time Estimate
[ ] Immediately
[ ] Within one week
[x] When possible
Types of changes
[x] Bug fix (non-breaking change which fixes an issue)
[ ] New feature (non-breaking change which adds functionality)
[ ] Breaking change (fix or feature that would cause existing functionality to change)
Checklist:
[ ] My change requires a change to the documentation.
Rebuild logic between partial missing observations and complete missing observation (free run case), and fixed some edge cases for the wishart Q.
Description
The free run is different than a normal run with partial missing observations for specific sets of sites. For the free run, we will estimate all state variables based on previous states and the potential Q from outside. For the normal run with partial missing observations, we will estimate state variables that match up with maximum types of observations across all sites. For the Wishart Q, some edge cases such as single site with single observation will be treated as Normal-Gamma conjugacy instead of MVN-Wishart conjugacy.
Motivation and Context
Review Time Estimate
Types of changes
Checklist: